[thesite] old links

[Daniel J. Cody]

On Tue, 19 Mar 2002, .jeff wrote:
> we are capturing them, but the redirect doesn't seem to be working properly in some browsers.
>
> cmd> GET /index.cfm?menu=8&cid=2396&catid=25 HTTP/1.0
> cmd> Host: www.evolt.org

'Host' is only needed in http 1.1, which your program should be using.
unless you're connecting to the server after you specify which file to
get, which seems wacky.

> i'm guessing most browsers don't correctly interpret the 301 error code, at least not ie6 or nn4.  nn6.2 performs the redirect properly.  nn4 just shows an ugly 404 page with a link to the article.  ie6 (and i suspect ie5.x also) just throw up an internal error page.  that's not terribly useful at all.  not sure what the solution is.

no, most browsers do interpret it correctly. if they're not, its because
friendly error messages are turned on(ie5x worked for me on PC and mac).
NN4 also showed an evolt page with a link and 'sorry' paragraph on os9.1,
linux and windows, so i'm not sure what you're seeing...

the problem, as demonstrated above, is that we're using the wrong
protocol version to test this out. in http 1.1 301 is a permanent move,
302 is 'found'. in http 1.0 301 is temporarily moved and 302 is perm.

try it from a telnet prompt or set your tools to use http1.1 and you'll
see

.djc.